Windows Server 2008 Configuration Network Load Balancing (service cluster)

When you recently configured SharePoint WFE, the customer mentioned that multiple WFE could load balance, so the next Network Load Balancing was studied.

When a server (including Web servers, FTP servers, or streaming media servers, etc.) is put into the network, as the number of clients increases, people often need servers that are more powerful and processing faster. To solve this problem, it is obviously not a good idea to replace the original server with a more powerful, faster server. But if you can add a new server to the processing power of the original server instead of replacing it, this scheme is undoubtedly more acceptable to the user. The presence of a Network Load Balancing (BALANCING,NLB) cluster is exactly what it does.

Working principle:

Both Node A and Node B have an externally used static IP address that creates an NLB cluster that generates a virtual IP that is in the same network segment as the IP of the node and that the externally advertised IP is a virtual cluster IP, and that the client is actually accessing node A or Node B.

Configuration steps:

1. Set the static IP in Node A and Node B, respectively:

2. Install NLB in Node A and Node B, respectively:

3. Open NLB on one of the nodes:

Click New Cluster:

Enter the IP of Node A, click on connect– "Next:

Add the cluster virtual IP, this virtual IP is the IP we want to access:

After the first node is configured, right-click on the newly created Cluster, select Add Host to Cluster, and join the second node. Once configured, the following should be done:

After aggregation is complete, the Web site is accessible through the cluster virtual IP:

You can see that this is 185 because it has a higher priority:

I have only one line on the index page of the 185 on 80 port: This is site on 185!

I have only one line on the index page of the 187 on 80 port: This is site on 187!

Detailed

1. Select the cluster operation mode (dual NIC unicast mode is recommended)

Unicast: Unicast mode refers to the network adapter of each node being re-assigned a virtual Mac (consisting of 02-BF and cluster IP addresses to ensure the uniqueness of this Mac). Because all of the network adapters that are bound to the cluster have the same Mac, there is no communication between the nodes in the case of a single NIC, which is one of the reasons why the dual NIC configuration is recommended. In order to avoid data flooding of the switch, it should be used in conjunction with VLANs.

Multicast: The network adapter also assigns a multicast MAC address shared by each node while preserving the original MAC address. Therefore, even though the nodes of a single NIC can communicate normally, most routes and switches do not support them very well.

IGMP multicast: IGMP multicast (which can only be selected when multicasting is selected), in addition to the benefits of inheriting multicast, NLB sends IGMP information every 60 seconds, so that multicast packets can only be sent to the correct switch port, avoiding the generation of switch data flooding.

2. Configure Port rules:

Cluster IP Address: Specifies the cluster IP to which the rule is directed

Port range: Default is all, you can specify the port range that the cluster listens to (for example, from 80 to 80, which means load balancing is only for Web services)

Protocol: Specifies the type of protocol that the cluster serves

Filter Mode:

A: Multiple hosts:

No similarity: The client's service requests are evenly distributed to every server within the cluster. Assume that there are 2 servers in the NLB cluster. When the request to the client, NLB will be the 1th request to the 1th server to process, the 2nd request to the 2nd server processing, 3rd requests to the 1th server to process, ... And so on Because all client-side opportunities are evenly distributed to each server, optimal load balancing can be achieved. If transaction processing is required, in order to be able to share the session state, the session state must be centrally stored in the states or database server, which is suitable for most applications.

Single similarity: A client's service request is pinned to a server within the cluster. When a request is received from a client, NLB decides which server to process according to the client's IP, that is, a server will only process requests from certain IPs. Because a service request for an IP is only fixed by one server, there is no problem with session state sharing, but it can cause load imbalance. This method is suitable for online communication protocols (such as FTP and PPTP) that need to support SSL set multiple online

Network (Class C): According to the IP Class C screen to decide which server to handle, that is, a server will only handle requests from some network segment c. This approach ensures that clients that use multiple proxies can be directed to the same server.

B. Single host: If you select this option, all requests within that port range will be handled by a single host, which will be used in conjunction with subsequent host precedence for host determination.

C. Disable this port range: Generally this option will be set in the port exception, that is, when we specify a larger range port, where there is one or several ports we do not need to be accessed by client users, we will use this rule to set up to prevent users from accessing this port request.

In addition, after the configuration is complete, the server will be restarted automatically, so there will be a short connection problem.

Windows Server 2008 Configuration Network Load Balancing (service cluster)